To test the psychometric properties of the Quality Indicator for Rehabilitative Care (QuIRC), refine the toolkit and compare results with service user experiences.

Following the initial development of the toolkit, it was translated into the languages of the partner countries and piloted. It was then refined to maximize a) its inter-rater reliability, b) its usability and c) its ability to deliver assessments relevant to each country's established systems of change at local, regional and national level. Managers of participating units were re-interviewed using the refined tool. QuIRC scores were compared against service users’ quality of life, autonomy, experiences of care and markers of recovery to assess whether the QuIRc could provide a proxy-assessment of the unit's promotion of service users’ autonomy and Recovery.

The tool was piloted in 20 units in each country (a total of 200 units). Inter-rater reliability was assessed using intra-class correlations and Cohen's Kappa coefficients. Factors with low reliability or extreme response biases were dropped. Remaining items were subjected to an exploratory factor analysis to test domain allocation and improve internal consistency. QuIRC domain ratings were compared by country and facility type and with service user assessments. The QuIRC was found to have high reliability, to be easy to use and there was high correlation between domain ratings and service users’ ratings.